XGBoost Predicts Major Cardiovascular Events in HFpEF and UA
Chinese multicenter cohort study (Jan 1, 2015–Dec 31, 2021) of 4,459 patients with coexisting HFpEF and unstable angina developed and externally validated machine-learning survival models to predict major adverse cardiovascular events (MACEs). A surv.xgboost.cox model using seven clinical and laboratory predictors achieved a C-index of 0.788 and 40-month AUC 0.807 in external validation, and was deployed as a public web risk calculator.
Key Points
- 1Develops surv.xgboost.cox survival model predicting 40-month MACEs in 4,459 HFpEF and UA patients
- 2Identifies seven predictors: diabetes, NT‑proBNP, SIRI, TyG‑BMI, triglycerides, platelets, atherogenic index
- 3Provides publicly available web calculator enabling clinicians to estimate individualized 40‑month MACE risks
Scoring Rationale
Strong multicenter, externally validated XGBoost survival model with web tool; limited generalizability beyond HFpEF and unstable-angina patients.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems